Notes on days of the week and other date-related aspects in three Greek inscriptions of the late Roman period This article consists of a series of comments, revisions, and new readings of three Greek inscriptions coming from different areas of the Roman Empire (Asia Minor, Thessaly, Gaul), and dating approximately from the third to the fifth century CE. I offer a restoration proposal and a new interpretation of inscription no. 1 – an votive dedication inscribed on an altar from Ankyra in Galatia – while my comments, revisions, and new readings of nos. 2 and 3 – two epitaphs, from Nea Anchialos in Thessaly and Augusta Treverorum in Gallia Belgica, respectively – concern primarily, though not exclusively, questions related to their date formulae.1 1) Inscribed altar, third century CE (?) (figs. 1–4) Square altar of pale limestone [63 x 59.5 (front) and 55 (shaft) cm. Letters height 2–2.5 cm.], broken at the top left and generally chipped and damaged. A deep basin has been cut into the top surface. Each face of the altar shows a different image in semi relief: below text (1) a thunderbolt (fig. 1); below text (2) a rounded item, interpreted by David French as a wreath (fig. 2); below text (3) a Nike holding a wreath (figs. 3–4); on the left face (4), a multi- petalled image, compass-drawn, circumscribed by a circle, interpreted by French as a flower 1 The present article results from the research I have been conducting since September 2013 as a member of the team working on the ERC-funded project Calendars in Antiquity and the Middle Ages: Standardization and Fixation. The project is based in the Department of Hebrew and Jewish Studies at UCL and is led by Sacha Stern. As part of this project, I have been looking at the origins and the process of diffusion and standardization of the seven- day week in the Roman Empire. To this end, I have carried out an extensive survey, collection, and study of references to days of the week and the seven-day week in the entire corpus of epigraphic, documentary, and literary sources in Greek and Latin coming from the territory encompassed by the Roman Empire. I have arranged these sources in the form of a database, which includes the inscriptions analysed in this article. I am grateful to Sacha Stern for reading a draft of this article and for his advice. I would also like to thank Georg Petzl for his helpful suggestions. None of them is responsible for remaining flaws. 1 (fig. 4). The inscription shows lunate letters (epsilon, sigma, and omega). The altar was found at Ankara, in Turkey, the site of ancient Ankyra in the region of Galatia; no further details on its provenance were recorded. It is now in the Roman Baths of Ankara (Location J III; Inv. no. 113.508.99). Editions: French 2003, 134, no. 33; SEG LIII 1439; Mitchell – French 2012, 385–387, no. 196. (1) Front face Ἑρµοῦ Διός (2) Right face Σελήνης Ἄρεως (3) Back face Ἀφροδείτη[ς] (4) Left face [Κρόνου Ἡλίου] L. 4: [Κρόνου Ἡλίου] supplevi. «(1) Of Hermes, of Zeus (2) Of Selene, of Ares (3) Of Aphrodite (4) [Of Kronos, of Helios]» In all previous editions of this inscribed altar the text on the left face (4) has been simply defined as ‘lost’ – no reconstruction proposals have been thus far suggested. While in the editio princeps French alludes solely to the fact that the altar belonged to ‘several gods’, in SEG Chaniotis points out that ‘the only common denominator of these divinities is that they have given their names to planets’, and concludes that ‘this may not be an altar, but an astronomical text’. As Mitchell and French later observed, however, the deep basin cut in the top surface, which was clearly designed to receive votive offerings, demonstrates that the stone was indeed used as an altar within the context of a cultic activity.2 My assumption is that this altar was dedicated to the seven gods of the planetary week. It should be observed that the order by which the names of the gods are inscribed on the four faces of the altar corresponds exactly to the order of the seven days in the 2 Mitchell – French 2012, 387. This further disproves, in my opinion, Arca’s argument (2005) that all altars with gods’ names in the genitive functioned as boundary stones of sanctuaries. Cf. Mitchell – French 2012, 386–387, who noted that Arca’s assumption ‘would lead to the conclusion that the present stone was a boundary stone of a sanctuary devoted to a very heterogeneous collection of named divinities’. 2 planetary week: if one starts reading the text from face (4) – as I restore it –, and follows with (2), (1), and (3), the resulting sequence of deities is Kronos, Helios, Selene, Ares, Hermes, Zeus, and Aphrodite, that is, in Roman terms, Saturn, Sol, Luna, Mars, Mercury, Jupiter, and Venus, i.e. to the planetary week sequence Saturday, Sunday, Monday, Tuesday, Wednesday, Thursday, and Friday. The explanation that is generally accepted for the order followed by the seven gods in the planetary week is conveniently illustrated by Cassius Dio (37.18–19), who takes up an astrological theory3 according to which the week was mapped out in one hundred and sixty-eight hours (7 days x 24 hours) with different influences ascribed to them. Dio informs us that the seven planetary gods, by order of distance from the Earth – Saturn, Jupiter, Mars, Sun, Venus, Mercury, and Moon – were assigned serially to the twenty-four hours of the day, and then to the one hundred and sixty-eight hours of the week, the god assigned to the first hour of each day also becoming the ‘lord’, or ‘governor’ of that particular day. Therefore each planet is assigned both to hours and to a whole day. According to this calculation the resulting sequence is indeed: day of Saturn (Saturday), day of the Sun (Sunday), day of the Moon (Monday), day of Mars (Tuesday), day of Mercury (Wednesday), day of Jupiter (Thursday), and day of Venus (Friday). This type of week starts on Saturday because Saturn is the furthest planet and therefore the first in the sequence of planets.4 Now my assumptions, i.e. that the left face of the altar was inscribed with the names of Kronos and Helios in the genitive, that the altar was dedicated to the seven gods of the 3 Vett. Val. 1.10. 4 That the planetary week started on Saturday is epigraphically confirmed by some of the earliest attestations of this temporal cycle, such as the Pompeian graffiti CIL IV 5202, CIL IV 6779, Inscr.It 13.2, 53, along with the Posillipo parapegma (Inscr.It. XIII.2, 52), which was attributed to the first century CE; as well as, in later periods, by a series of inscribed and illustrated parapegmata (e.g. Inscr.It XIII.2, 56 – the Trajan’s Bath parapegma, dated to the Constantinian period –; Rostovtzeff 1936, 40–46 no. 622 – the Dura Europos parapegma of c. 165–257 CE), by inscribed objects such as a golden bracelet from Syria (de Witte 1877, 83–84, pl. 8.4–5, who ascribed it to c. 300 CE), a limestone cornice from ancient Thuburbo Maius in Africa Proconsularis (ILPBardo 346, undated), and further epigraphic and artifactual evidence. 3 planetary week, and that the order of the deities’ names on it corresponds to the order of the seven days in the planetary week, which indeed began on Saturday, are supported by the fact that Aphrodite’s name appears on its own on one face of the altar.5 Obviously, seven god-names cannot be distributed evenly onto the four vertical faces of a square altar; if the week begins on Saturday, and two deities are inscribed on each side, then for the fourth and last side of the altar there is only one deity left, i.e. Aphrodite. As for the date of the altar, in his 2003 edition French attributed it tentatively to the first or second century CE, without providing any explanation for his supposition (the same chronological reference can be found in SEG LIII 1439, with question mark); in the second edition, Mitchell – French dated the inscribed altar to the third century on the basis of the letter forms (essentially the lunate letters) and the style of the reliefs that adorn the four vertical faces. The later date would fit better in the context of the diffusion of the seven- day, planetary week in the Roman world: the ancient sources indeed show that the planetary week started enjoying great popularity during the third century CE.6 5 The restoration [Κρόνου Ἡλίου] is further substantiated by the relief appearing on that face of the altar: as pointed out to me by Georg Petzl, the multi-petalled image circumscribed by a circle, which was previously interpreted as a flower, might instead symbolise the Sun (Fig. 4). In fact, as Petzl noted, each face of the altar shows a relief that appears to match the text inscribed above it: the thunderstorm under Ἑρµοῦ Διός clearly relates to Zeus (Fig. 1); the Nike holding a wreath (Figs. 3–4) could refer to Aphrodite (cf. Verg. ecl. 10.69: omnia vincit Amor); the rounded object under Σελήνης Ἄρεως (Fig. 2) represents a moon crescent, rather than a wreath, and is therefore associated with Σελήνη. A close parallel for this manner of representing a lunar crescent appears, e.g., in Petzl 1994, 81, no. 63. 6 Apparently starting with the reign of the Severans: cf., e.g., Cassius Dio (37.18), who writes at the time of Septimius Severus and says that the practice of naming the days of the week after the seven planets was then general not only at Rome, but ‘among all mankind’.